Patient-specific PTV margins for liver stereotactic body radiation therapy determined using support vector classification with an early warning system for margin adaptation

Artificial intelligence-based motion tracking in cancer radiotherapy: A review

Radiotherapy aims to deliver a prescribed dose to the tumor while sparing neighboring organs at risk (OARs). Increasingly complex treatment techniques such as volumetric modulated arc therapy (VMAT), stereotactic radiosurgery (SRS), stereotactic body radiotherapy (SBRT), and proton therapy have been developed to deliver doses more precisely to the target. While such technologies have improved dose delivery, the implementation of intra-fraction motion management to verify tumor position at the time of treatment has become increasingly relevant. Artificial intelligence (AI) has recently demonstrated great potential for real-time tracking of tumors during treatment. However, AI-based motion management faces several challenges, including bias in training data, poor transparency, difficult data collection, complex workflows and quality assurance, and limited sample sizes. This review presents the AI algorithms used for chest, abdomen, and pelvic tumor motion management/tracking for radiotherapy and provides a literature summary on the topic. We will also discuss the limitations of these AI-based studies and propose potential improvements.

Commissioning of a solid tank design for fan-beam optical CT based 3D radiation dosimetry

Objective.Optical computed tomography (CT) is one of the leading modalities for imaging gel dosimeters used in the verification of complex radiotherapy treatments. In previous work, a novel fan-beam optical CT scanner design was proposed that could significantly reduce the volume of the refractive index baths that are commonly found in optical CT systems. Here, the proposed scanner has been manufactured and commissioned.Approach.Image reconstruction is performed through algebraic reconstruction technique and iterated using the fast iterative shrinkage-thresholding algorithm (FISTA) algorithm. Ray tracing for algebraic reconstruction was performed using an in-house developed ray tracing simulator. A set of Sylgard® 184 phantoms were created to commission spatial resolution, geometric deformity, contrast-to-noise ratio (CNR), and scan settings.Main Results.The scanner is capable of a 0.929 mm-1spatial resolution, observed at 200 iterations, although the spatial resolution is highly dependent on the number of iterations. The geometric distortion, measured by scanning a needle phantom with the prototype scanner as well as a conventional x-ray CT was found to be within <0.25 mm. The CNR was found to peak between 65 and 190 occurring between 50 and 100 iterations and was highly dependent on the region chosen for background noise calculation. The proposed scanner is capable of scanning and reading out slices in less than 1 min per slice.Significance.This work displays the viability of a fan-beam optical CT scanner with minimal index matching using ray-traced algebraic reconstruction.

Comprehensive Treatment Uncertainty Analysis and PTV Margin Estimation for Fiducial Tracking in Robotic Liver Stereotactic Body Radiation Therapy

OBJECTIVE

This study aims to quantify the uncertainties of CyberKnife Synchrony fiducial tracking for liver stereotactic body radiation therapy (SBRT) cases, and evaluate the required planning target volume (PTV) margins.

METHODS

A total of 11 liver tumor patients with a total of 57 fractions, who underwent SBRT with synchronous fiducial tracking, were enrolled for the present study. The correlation/prediction model error, geometric error, and beam targeting error were quantified to determine the patient-level and fraction-level individual composite treatment uncertainties. The composite uncertainties and multiple margin recipes were compared for scenarios with and without rotation correction during treatment.

RESULTS

The correlation model error-related uncertainty was 4.3±1.8, 1.4±0.5 and 1.8±0.7 mm in the superior-inferior (SI), left-right, and anterior-posterior directions, respectively. These were the primary contributors among all uncertainty sources. The geometric error significantly increased for treatments without rotation correction. The fraction-level composite uncertainties had a long tail distribution. Furthermore, the generally used 5-mm isotropic margin covered all uncertainties in the left-right and anterior-posterior directions, and only 75% of uncertainties in the SI direction. In order to cover 90% of uncertainties in the SI direction, an 8-mm margin would be needed. For scenarios without rotation correction, additional safety margins should be added, especially in the superior-inferior and anterior-posterior directions.

CONCLUSION

The present study revealed that the correlation model error contributes to most of the uncertainties in the results. Most patients/fractions can be covered by a 5-mm margin. Patients with large treatment uncertainties might need a patient-specific margin.

Machine Learning Radiomics Model for External and Internal Respiratory Motion Correlation Prediction in Lung Tumor

Objectives: The complexity and specificity of lung tumor motion render it necessary to determine the external and internal correlation individually before applying indirect tumor tracking. However, the correlation cannot be determined from patient respiratory and tumor clinical characteristics before treatment. The purpose of this study is to present a machine learning model for an external/internal correlation prediction that is based on computed tomography (CT) radiomic features. Methods: 4-dimensional computed tomography (4DCT) images of 67 patients were collected retrospectively, and the external/internal correlation of lung tumor was calculated based on Spearman's rank correlation coefficient. Radiomic features were extracted from average intensity projection and the light gradient boosting machine (LightGBM)-based cross-validation (the recursive elimination method) was used for feature selection. The LightGBM framework forecasting models with classification thresholds 0.7, 0.8, and 0.9 are established using stratified 5-fold cross-validation. Model performance was assessed using receiver operating characteristics, sensitivity, and specificity. Results: There were 16, 18, and 13 features selected for models 0.7, 0.8, and 0.9, respectively. Texture features are of great importance in external/internal correlation prediction compared to other features in all models. The sensitivities of the predictions in models 0.7, 0.8, and 0.9 were 0.800 ± 0.126, 0.829 ± 0.140, and 0.864 ± 0.086, respectively. The specificities were 0.771 ± 0.114, 0.936 ± 0.0581, and 0.839 ± 0.101, whereas the area under the curve (AUC) was 0.837, 0.946, and 0.877, respectively. Conclusions: Our findings indicate that radiomics is an effective tool for respiratory motion correlation prediction, which can extract tumor motion characteristics. We proposed a machine learning framework for correlation prediction in the motion management strategy for lung tumor patients.

Applications of artificial intelligence in stereotactic body radiation therapy

This topical review focuses on the applications of artificial intelligence (AI) tools to stereotactic body radiation therapy (SBRT). The high dose per fraction and the limited number of fractions in SBRT require stricter accuracy than standard radiation therapy. The intent of this review is to describe the development and evaluate the possible benefit of AI tools integration into the radiation oncology workflow for SBRT automation. The selected papers were subdivided into four sections, representative of the whole radiotherapy process: ‘AI in SBRT target and organs at risk contouring’, ‘AI in SBRT planning’, ‘AI during the SBRT delivery’, and ‘AI for outcome prediction after SBRT’. Each section summarises the challenges, as well as limits and needs for improvement to achieve better integration of AI tools in the clinical workflow.

Novel Clinically Weight-Optimized Dynamic Conformal Arcs (WO-DCA) for Liver SBRT: A Comparison with Volumetric Modulated Arc Therapy (VMAT)

PURPOSE

To evaluate the feasibility of shortening the duration of liver stereotactic radiotherapy (SBRT) without jeopardizing dosimetry or conformity by utilizing weight-optimized dynamic conformal arcs (WO-DCA) as opposed to volumetric modulated arc therapy (VMAT) for tumors away from critical structures.

METHODS

Nineteen patients with liver metastasis were included, previously treated with 50 Gy in 4 fractions with VMAT technique using two partial coplanar arcs of 6 MV beams delivered in high-definition multi-leaf collimator (HD-MLC). Two coplanar partial WO-DCA were generated on Pinnacle treatment planning system (TPS) for each patient; and MLC aperture around the planning target volume (PTV) was automatically generated at different margins for both arcs and maintained dynamically around the target during arc rotation. Weight of the two arcs using optimization method was adjusted between the arcs to maximize tumor coverage and protect organs at risk (OAR) based on the RTOG-0438 protocol.

RESULTS

The WO-DCA plans successfully "agreed" with the standard VMAT for OAR (liver, spinal cord, stomach, duodenum, small bowel, and heart) and PTV (Dmean, D98%, D2%, CI, and GI), with superior mean quality assurance (QA) pass rate (97.06 vs 93.00 for VMAT; P < 0.001 and t = 8.87). Similarly, the WO-DCA technique additionally reduced the beam-on time (3.26 vs 4.43; P < 0.001) and monitor unit (1860 vs 2705 for VMAT; P < 0.001) values significantly.

CONCLUSION

The WO-DCA plans might minimize small-field dosimetry errors and defeat patient-specific VMAT QA requirements due to the omission of MLC beam modulation through the target volume. The WO-DCA plans may additionally enable faster treatment delivery times and lower OAR without sacrificing target doses in SBRT of liver tumors away from critical structures.